Search

Claude Code skills that make Claude (Fable) the conductor of AI worker fleets: many parallel OpenAI Codex workers plus Opus design agents, organized into campaigns, waves, squads, and review gates.

Fable 指揮官工作流 — 最強模型只做規劃/審查，研究與執行派給你選的模型；maker/verifier 分離的執行迴圈。Commander workflow for Claude Code: plan & review on the top model, delegate research/execution, based on Loop Engineering. 中英雙語。

Six Claude Code skills that harden Opus 4.8 toward frontier behavior — written by Fable 5, pressure-tested on the target model with transcripts included.

OpenAI Realtime API (gpt-realtime-2) integration for Unoverse platform

Reserved-namespace placeholder for TUNLstack, the orchestration layer for Claude Code (a VS Code extension at https://tunlstack.com).

No description available

This template should help get you started developing with Vue 3 in Vite.

OpenAI-compatible agent-core/v1 provider adapter

A personal card for Rabyâ Raghib (@rabraghib)

Guest-side bindings for the Tauri scoped storage plugin

llms.txt cache-first tool — fetch and cache llms.txt from any domain, inject guidance so the model uses it before deep-fetching docs

Storage adapter for Firebase Storage

DuckDB-based reader for an options-chain data lake stored in Azure Data Lake Storage

Calibrate -> Optimize -> Gate -> Serve: automatic prompt optimization with a calibrated LLM judge, statistical deploy gating, and a versioned serving registry.

Cloudflare D1 state store for @tokenflight/fiat

A terminal chat client for local LLMs — Ollama, LM Studio, MLX, and any OpenAI-compatible server.

LCAP 前端扩展组件库 AI Skill（Claude Code / Cursor / Codex）

面向 AI 架构创新的数学研究操作系统 v3：三层正交架构（思想透镜 / 数学知识库 / 设计翻译层），自动诊断用户意图并路由。Math Research OS for AI architecture design: thinking lenses + math knowledge base + design translation layer with auto intent diagnosis.

鸿蒙三方库适配统一入口（高效流程版）— 自动判定 Flutter 插件 / React Native 模块 / 原生 Android 库，路由到对应 agent 完成 HarmonyOS 适配。安装到 OpenCode / ClaudeCode 宿主。

Optional OpenClaw pre-model media guard for obsidian-cli-plugins that stages uploads without spending LLM tokens.